Back to projects
Project

Agentic AI Technical Sales Consultant (Web)

A domain-aware website assistant that guides buyers through process selection, material constraints, and feasibility trade-offs — then captures an RFQ-quality summary for sales engineering review.

Industrial manufacturing plant with AI dashboards
Guided feasibility to RFQ-ready summary — the buyer journey at a glance.

Implementation-focused summary

A website that behaves like a technical sales engineer

Designed for complex manufacturing buyers who need clear process recommendations and constraints — not generic chat responses.

Problem

Visitors arrived with feasibility questions (process selection, geometry/volume constraints, material trade-offs). Traditional forms and email-based intake created high bounce rates, repetitive engineer effort, and incomplete RFQs requiring back-and-forth.

Solution

An agentic AI consultant embedded into the website that gathers requirements, recommends one or more manufacturing processes with trade-offs, and produces a structured RFQ summary aligned with CRM fields.

How it works

1

Guidance

Process selection + constraints + trade-offs in clear, buyer-friendly language.

2

Knowledge

Retrieval anchored on company capabilities, tolerances, materials, and past projects.

3

RFQ capture

Structured intake and summarization for fast sales engineering review.

Architecture & Stack

  • LLM agent with tool orchestration for capability lookup, process recommendations, and RFQ summarization.
  • Curated knowledge base indexed for retrieval (processes, tolerances, materials, constraints).
  • Frontend web chat interface backed by a scalable API layer.
  • Outputs mapped to standardized CRM/RFQ fields for consistent lead qualification.

Impact

  • Improved inquiry quality by capturing complete requirements in one session.
  • Reduced repetitive engineering time on early-stage education and feasibility questions.
  • Higher conversion by providing interactive guidance instead of static forms.

Your role

Designed the knowledge representation, built the agent orchestration and tools, integrated into the website, and tuned the experience based on real user transcripts.